[PATCH] sched: consider WF_SYNC to find idle siblings

[Andrei Vagin <avagin@google.com>]

From: Andrei Vagin <avagin@gmail.com>

WF_SYNC means that the waker goes to sleep after wakeup, so the current
cpu can be considered idle if the waker is the only process that is
running on it.

The perf pipe benchmark shows that this change reduces the average time
per operation from 8.8 usecs/op to 3.7 usecs/op.

Before:
 $ ./tools/perf/perf bench sched pipe
 # Running 'sched/pipe' benchmark:
 # Executed 1000000 pipe operations between two processes

     Total time: 8.813 [sec]

       8.813985 usecs/op
         113456 ops/sec

After:
 $ ./tools/perf/perf bench sched pipe
 # Running 'sched/pipe' benchmark:
 # Executed 1000000 pipe operations between two processes

     Total time: 3.743 [sec]

       3.743971 usecs/op
         267096 ops/sec

Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Ben Segall <bsegall@google.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Daniel Bristot de Oliveira <bristot@redhat.com>
Cc: Valentin Schneider <vschneid@redhat.com>
Signed-off-by: Andrei Vagin <avagin@gmail.com>
---
 kernel/sched/fair.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index e4a0b8bd941c..40ac3cc68f5b 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -7245,7 +7245,8 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int wake_flags)
 		new_cpu = find_idlest_cpu(sd, p, cpu, prev_cpu, sd_flag);
 	} else if (wake_flags & WF_TTWU) { /* XXX always ? */
 		/* Fast path */
-		new_cpu = select_idle_sibling(p, prev_cpu, new_cpu);
+		if (!sync || cpu != new_cpu || this_rq()->nr_running != 1)
+			new_cpu = select_idle_sibling(p, prev_cpu, new_cpu);
 	}
 	rcu_read_unlock();
 
-- 
2.34.1